Saved in:
Bibliographic Details
Main Authors: Shaska, Joni, Mitra, Urbashi
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2507.21570
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918107033894912
author Shaska, Joni
Mitra, Urbashi
author_facet Shaska, Joni
Mitra, Urbashi
contents This paper proposes a novel framework for causal discovery with asymmetric error control, called Neyman-Pearson causal discovery. Despite the importance of applications where different types of edge errors may have different importance, current state-of-the-art causal discovery algorithms do not differentiate between the types of edge errors, nor provide any finite-sample guarantees on the edge errors. Hence, this framework seeks to minimize one type of error while keeping the other below a user-specified tolerance level. Using techniques from information theory, fundamental performance limits are found, characterized by the Rényi divergence, for Neyman-Pearson causal discovery. Furthermore, a causal discovery algorithm is introduced for the case of linear additive Gaussian noise models, called epsilon-CUT, that provides finite-sample guarantees on the false positive rate, while staying competitive with state-of-the-art methods.
format Preprint
id arxiv_https___arxiv_org_abs_2507_21570
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Causal Link Discovery with Unequal Edge Error Tolerance
Shaska, Joni
Mitra, Urbashi
Signal Processing
This paper proposes a novel framework for causal discovery with asymmetric error control, called Neyman-Pearson causal discovery. Despite the importance of applications where different types of edge errors may have different importance, current state-of-the-art causal discovery algorithms do not differentiate between the types of edge errors, nor provide any finite-sample guarantees on the edge errors. Hence, this framework seeks to minimize one type of error while keeping the other below a user-specified tolerance level. Using techniques from information theory, fundamental performance limits are found, characterized by the Rényi divergence, for Neyman-Pearson causal discovery. Furthermore, a causal discovery algorithm is introduced for the case of linear additive Gaussian noise models, called epsilon-CUT, that provides finite-sample guarantees on the false positive rate, while staying competitive with state-of-the-art methods.
title Causal Link Discovery with Unequal Edge Error Tolerance
topic Signal Processing
url https://arxiv.org/abs/2507.21570